Baseflow Ratio in Catchments with Regolith-Dominated Groundwater Circulation of Different Lithology—Comparison of Kille’s, Rambert’s and Hydrograph Separation Methods

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors present research on the separation of hydrographs in complex lithological/geological sites. Three models are presented. The idea is very promising; such problems are important, especially given the challenges of flooding, but also for drought analysis.

The authors present the problems in the real case study, which is always promising and adds value to the overall research. Unfortunately, the realization and the entire presentation of the research are not at the required level.

-The literature review is poorly written. There is a lack of literature review that presents the separation of hydrographs in complex lithological/geological sites. Also, authors cannot cite 10 or more references at the end of a sentence and expect readers to look up each reference to find details about the analyzed part of existing knowledge.

-It is not elaborated why these three methods for the separation of the hydrographs are selected. Please check the above comments. In other words, methodology is poorly defined.

-Only the topographic map of the analyzed site is not enough. There is a lack of vertical stratigraphy profiles. All conclusions drawn from such a limited input lithological structure are superficial. Subsurface and surface flows are not described here, so because of this, I insist on these profiles.

-The authors performed a separation of the hydrograph. In the final conclusion, there is no real, well-defined conclusion. Figures 6-9 are presented, but there are no real answers to the question of which model is better. It is not such an easy task, but it should be, and it could be provided. 

-In order to get answers to the previous questions, I am proposing a multicriteria method for solving this issue. Promethee or some other method will be quite fine, because here the criteria and importance of the particular method can be easily defined.

 

Author Response

Comments 1: The literature review is poorly written. There is a lack of literature review that presents the separation of hydrographs in complex lithological/geological sites. Also, authors cannot cite 10 or more references at the end of a sentence and expect readers to look up each reference to find details about the analyzed part of existing knowledge.

Response 1: Thank you for your comment. Regarding the citations in the introduction, we tried to follow the instructions mentioned in the template for manuscripts, that it should be brief and that all the citations mentioned in a sentence should be marked with numbers in brackets before punctuation. Although we agree, that the overview is missing names and works of the cited authors, we simply thought that this was the proper way to summarize works in this journal.
After your review, we revised the whole literature overview to the best of our knowledge and added missing parts of the citations. The changes to the literature overview can be found in the Introduction section of the revised manuscript from line No. 37 on page No. 1 to line No. 122 on page No. 3. The References section was adjusted accordingly.

Comments 2: It is not elaborated why these three methods for the separation of the hydrographs are selected. Please check the above comments. In other words, methodology is poorly defined.

Response 2: Thank you for your revision of methodology. We agree that it was missing the above-mentioned explanation, therefore this was addressed and clarified in the manuscript. You can find the added explanation on page No. 11 in two paragraphs from lines No. 323 to 343.

Comments 3: Only the topographic map of the analyzed site is not enough. There is a lack of vertical stratigraphy profiles. All conclusions drawn from such a limited input lithological structure are superficial. Subsurface and surface flows are not described here, so because of this, I insist on these profiles.

Response 3: Thank you for your comment. We agree with this and sufficient changes were made in the manuscript accordingly. The added litostratigraphical profiles for each unit can be found in the Materials and Methods section of the revised manuscript. On pages No. 8, 9 and 10 as Figures No. 4, 5 and 6, with explanations in the text on page No. 7 (lines 216-220) for Crystalline HG unit, on pages No. 7 & 8 (lines 236 to 259) for Flysch zone HG unit and on pages No. 9 (lines 274-283) for Inner Carpathian Paleogene HG unit.

Comments 4: The authors performed a separation of the hydrograph. In the final conclusion, there is no real, well-defined conclusion. Figures 6-9 are presented, but there are no real answers to the question of which model is better. It is not such an easy task, but it should be, and it could be provided.

Response 4: Thank you for your observation. At first, we did not really wanted to „pick“ or rank the best or worst method, we simply wanted to present an overview of the comparison and the differences, but we agree that there is the potential for providing such information at least in relative terms of methods’ applicability. Such paragraph with our summary for usability of methods was added to the conclusions of the manuscript after your review. The corresponding lines No. 717- 740 are on page No. 24 of the revised manuscript.

Comments 5: In order to get answers to the previous questions, I am proposing a multicriteria method for solving this issue. Promethee or some other method will be quite fine, because here the criteria and importance of the particular method can be easily defined.

Response 5: Thank you for your suggestion. We decided in this case, not to use a multicriteria method to simply stating which method is the best. Each method may be more applicable in different conditions and we think, that we summarized this adequately in the revised conclusions after your above-mentioned comment. We may look for this type of analysis in future of our research though. Thank you again for inspiration.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The analysis of recession curves in hydrographs and their hydrogeological interpretation is an important part of the content of all general hydrology texts, which typically present some of the most widely used methodologies in this regard.

This manuscript compares the results of three such methodologies in 42 small basins in Slovakia. The climate is assumed to be the same, although there is significant variation in average altitude (which in turn influences the amount of precipitation) and in topography. The key variable considered in each basin is the relative importance of groundwater discharge (baseflow) compared to the total discharge. The manuscript compares the results obtained in each basin using each of the methods. This reveals similarities and differences that are discussed, which can be considered the manuscript's main methodological value. In this respect, there is no comment on whether this type of comparative analysis has been previously carried out in other parts of the world or whether this is the first time the authors are aware of it having been done, which would add more interest to the discussion of the results. The methods used are likely those commonly employed in Slovakia (and perhaps in other countries as well), but they are unfamiliar to me. Therefore, I focused on the section of the manuscript where the fundamentals of these methods were presented, but I was disappointed by the limited information provided. I expected them to be published in easily accessible international journals written in English, but to my surprise, this is not the case. In fact, although I cannot be certain, I believe these methods are not mentioned in the review on this topic in a 1995 article in the Journal of Hydrology cited in the manuscript (reference 7).

Essentially, the manuscript invokes, by way of a numerical index, the more or less ruggedness character of the terrain of each basin to explain spatial differences, as previously mentioned. The only considerations regarding the potential influence of aquifer type on the key variable studied are addressed in the manuscript by grouping the basins into three supposedly different groups, corresponding to the regolith (several tens of meters thick in all cases) formed primarily on: a) crystalline rocks, b) folded Paleogene flysch, and c) unfolded Paleogene flysch. However, no lithological criteria are provided to support this distinction between the three regolith groups, particularly regarding predominant grain size, soil texture, or any other characteristic that might determine permeability or infiltration capacity in these surface formations. Nor is anything mentioned about groundwater flow systems, suggesting (although not explicitly stating it) the existence in all watersheds of a relatively simplistic scheme in which the drainage of the regolith occurs upstream of the flow measurement stations and that the boundaries of the groundwater flow systems coincide, equally in all cases, with the boundaries of the studied watersheds. These aspects severely limit the hydrological value of the study's results.

In short, in its current state, the manuscript has to my mind little methodological or hydrological value, and I therefore recommend improvements in the aspects mentioned. Otherwise, the manuscript is easy to understand (although I am not a native English speaker) and is well-structured and illustrated. The bibliography seems relevant, with the aforementioned exceptions regarding works not published in English.

Author Response

Comments 1: There is no comment on whether this type of comparative analysis has been previously carried out in other parts of the world or whether this is the first time the authors are aware of it having been done, which would add more interest to the discussion of the results.

Response 1: Thank you for your comment. Although we did not see it as important stating it in the text of the manuscript, a study dedicated to these three specific methods used in these specific conditions does not exist, to our knowledge. 
After the reviews, we rebuild the literature overview in the Introduction of the manuscript, and studies that are similar, to any extent, we indeed provided citations to the best of our knowledge. The revised literature overview can be found in the Introduction section of the revised manuscript on pages No. 1 to 3, lines No. 37 to 122. The Reference section was adjusted accordingly.

Comments 2: The methods used are likely those commonly employed in Slovakia (and perhaps in other countries as well), but they are unfamiliar to me. Therefore, I focused on the section of the manuscript where the fundamentals of these methods were presented, but I was disappointed by the limited information provided. I expected them to be published in easily accessible international journals written in English, but to my surprise, this is not the case.

Response 1: Indeed, these methods were proven as satisfactory in Slovakia in the past, which is one of the factors why these specific methods were used. There were partial studies carried out in the past and published (mostly locally in Slovak language, although usually with english abstract and summary included).
One of our goals was to expand on these partial studies to more overall extent for the territory of Western Carpathians. Due to this, we focused more on the comparative element of the study and the bases of these methods were described perhaps a little vaguely. After your comment, we significantly elaborated and expanded the description of methodologies used in the manuscript. Thank you again for such in depth review of the methodology. You can find the added descriptions in the Materials and Methods section of the revised manuscript on pages No. 11 to 13 in lines No. 344 to 396 supplemented also by Figures No.7 & 9.

Comments 3: I believe these methods are not mentioned in the review on this topic in a 1995 article in the Journal of Hydrology cited in the manuscript (reference 7).

Response 3: We assume you meant refference number [6] (original manuscript) - Tallaksen, 1995. This was indeed a mistake on our part, as this was supposed to be a different Tallaksen refference - a book by Tallaksen and Van Lanen, 2004: Hydrological Drought – Processes and Estimation Methods for Streamflow and Groundwater, which describes baseflow separation methods such as BFI. But this is only one of the methods used in this study. The other two methods are different and are mentioned in other parts of the manuscript with different refferences: numbers [2 and 11] (original manuscript) for Kille’s method and numbers [8 and 15] (original manuscript) for Rambert’s method.
Thank you for your comment and review of the literature, this mistake was corrected in the Introduction section of the revised manuscript on page No. 2 lines 57 and 58, as well as in the References section on page No. 25 line No. 799.

Comments 4: The only considerations regarding the potential influence of aquifer type on the key variable studied are addressed in the manuscript by grouping the basins into three supposedly different groups, corresponding to the regolith (several tens of meters thick in all cases) formed primarily on: a) crystalline rocks, b) folded Paleogene flysch, and c) unfolded Paleogene flysch. However, no lithological criteria are provided to support this distinction between the three regolith groups, particularly regarding predominant grain size, soil texture, or any other characteristic that might determine permeability.

Response 4: Thank you for your comment. After your review, the basic geological parameters of the studied regolith groups were addressed and supplemented, in form of lithostratigraphic profiles and their respective text descriptions in the Materials and Methods section of the manuscript. The added profiles for each unit can be found on pages No. 7, 8 and 9 as Figures No. 4, 5 and 6, with explanations in the text on page No. 7 (lines 216-220) for crystalline HG unit, on pages No. 7 & 8 (lines 236 to 259) for flysch zone HG unit and on pages No. 9 (lines 274-283) for Inner Carpathian Paleogene HG unit.
Also, basic hydraulic parameters which are related to permeability were added in the manuscript near the above-mentioned paragraphs. You can find them on pages No. 7 (lines 224-228) for crystalline HG unit; page No. 9 (lines 263-273) for flysch zone HG unit; and page No. 10 (lines 287-290) for Inner Carpathian Paleogene HG unit.

Comments 5: "... or infiltration capacity in these surface formations."

Response 5: Thank you for your observation. To determine, or estimate the relative infiltration capacity from the unevaporated precipitation point of view was one of the goals of this study as stated in the Conclusions of the manuscript (lines No. 677 to 692). This is closely related to your next comment which is addressed below.

Comments 6: Nor is anything mentioned about groundwater flow systems, suggesting (although not explicitly stating it) the existence in all watersheds of a relatively simplistic scheme in which the drainage of the regolith occurs upstream of the flow measurement stations and that the boundaries of the groundwater flow systems coincide, equally in all cases, with the boundaries of the studied watersheds.

Response 6: Yes indeed, the general concept and factor for selecting these headwater catchments was that the morphologic boundaries correspond with the hydrogeological boundaries, and therefore the drainage of the regolith zone occurs solely in the upstream zone of the watershed (in direction from the water gauging station), while eliminating any lateral cross-boundary flow. This concept is as much simplified as it needs to be for this kind of research, while still taking into account the scientific bases of surface and groundwater runoff. 
Thank you for your comment, after your review we implemented this explanation to the manuscript. The addition can be found on page No. 4, in the Materials and Methods section of the revised manuscript, with corresponding lines from No. 151 to 158. And also on page No. 10 in the same section in lines from No. 299 to 302.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper compares baseflow ratios across 42 Western Carpathian catchments using three methods—Kille's statistical approach, the Local Minimum (BFI) filter, and Rambert's recession-curve analysis—across crystalline, folded flysch, and unfolded Paleogene lithologies with regolith-dominated groundwater circulation. Its main contribution is a systematic, multi-method assessment demonstrating that crystalline catchments yield the highest baseflow ratios while folded flysch shows the lowest, with Rambert's method consistently overestimating by 10–12%. Strengths include long-term discharge records, clear geological stratification, and quantitative inter-method correlation analysis, offering valuable regional benchmarks for hydrogeological surveys in similar Central European mountain terrains.

General Comments

1. Although the introduction mentions that multiple methods have been "applied simultaneously for comparison" (Paragraph 2, Refs. 19–25), it fails to clearly articulate the essential distinction between this study and existing comparative studies.
2. The time step length of N = 20 days is mentioned, yet no explanation is provided as to how this parameter was "established as optimal" (Lines 187–188), lacking either a citation or a sensitivity analysis.
3. The authors observe that crystalline catchments exhibit the highest baseflow ratios (44–56%) and folded flysch the lowest (29–37%), but the discussion remains at the level of descriptive statistical associations (elevation, terrain ruggedness index). It does not delve deeper into explaining: why crystalline rocks, with well-developed fractures, show higher baseflow ratios (fractured rock masses are typically highly conductive but weakly storative); or how tectonic deformation in folded flysch specifically affects groundwater recharge pathways.
4. Rambert's method yields anomalously high baseflow ratios at several stations, with a standard deviation markedly larger than the other two methods (Table 7). The authors attribute this to "the subjectivity of recession-curve interpretation" (Lines 460–463), yet fail to explain: whether high values result from the inclusion of non-baseflow components (e.g., interflow) during recession-curve fitting; or whether low values arise from excessively rapid recession leading to underestimation of the baseflow component.
5. The conclusions are excessively long, largely repeating numerical values from the Results section (e.g., specific ranges for each method across lithologies) without adequate synthesis or refinement.
6. The language quality requires improvement; careful proofreading of grammar, tense consistency, and logic is recommended.

Specific Comments

7. Figure 3 displays a ranked-events curve, but the specific algorithmic steps for extracting baseflow values from this curve are not explained.
8. It is recommended to add the standard deviation or coefficient of variation for the multi-year average baseflow ratio at each station to Tables 4–6.
9. Figure 1: The watershed boundary lines and administrative boundary lines employ similar line styles, making them difficult to distinguish.
10. Table 2 appears twice (on Pages 4 and 5).

Author Response

Comments 1: Although the introduction mentions that multiple methods have been "applied simultaneously for comparison" (Paragraph 2, Refs. 19–25), it fails to clearly articulate the essential distinction between this study and existing comparative studies.

Response 1: Thank you for your observation and comment. The shortcomings of the Introduction section were addressed and corrected to the best of our knowledge based on the reviewer reports provided. Changes to the literature overview can be found in the Introduction section of the revised manuscript, from line No. 37 on page No. 1 to line No. 122 on page No. 3. The Reference section was adjusted accordingly.

Comments 2: The time step length of N = 20 days is mentioned, yet no explanation is provided as to how this parameter was "established as optimal" (Lines 187–188), lacking either a citation or a sensitivity analysis.

Response 2: Thank you for your comment, this observation is much appreciated. From our perspective it was clear and sufficiently substainted (with only numerical reference provided), but thanks to your comment we see that for readers it may be less obvious and needs to be specified in more detail. This explanation can be seen in the Methods and Materials section of the revised manuscript on pageNo. 13 lines 391 to 394.

Comments 3:  The authors observe that crystalline catchments exhibit the highest baseflow ratios (44–56%) and folded flysch the lowest (29–37%), but the discussion remains at the level of descriptive statistical associations (elevation, terrain ruggedness index). It does not delve deeper into explaining: why crystalline rocks, with well-developed fractures, show higher baseflow ratios (fractured rock masses are typically highly conductive but weakly storative); or how tectonic deformation in folded flysch specifically affects groundwater recharge pathways.

Response 3: Thank you for your comment on the Discussion of the results in the manuscript. We agree, that it is missing a deeper explanation of the more hydrogeological aspects affecting the ratio of baseflow to total streamflow for studied HG units. After your review, we added a paragraph dedicated to explaining this in more detail on page No. 20 (lines 585 to 597). However, we still think that the morphologic parameters remain significant factor for baseflow generation in individual catchments and go hand in hand with other factors.

Comments 4: Rambert's method yields anomalously high baseflow ratios at several stations, with a standard deviation markedly larger than the other two methods (Table 7). The authors attribute this to "the subjectivity of recession-curve interpretation" (Lines 460–463), yet fail to explain: whether high values result from the inclusion of non-baseflow components (e.g., interflow) during recession-curve fitting; or whether low values arise from excessively rapid recession leading to underestimation of the baseflow component.

Response 4: Thank you for your comment and this observation. This is indeed true, as a hybrid genetic algorithm (embdded within the RC 4.0 software) was used to create MRCs, which could possibly cause this anomalous high or low baseflow ratios within the catchments. After your review, this issue was addressed also in the Discussion section of the revised manuscript. The provided explanation should be seen on page No. 22, lines 644-652.

Comments 5: The conclusions are excessively long, largely repeating numerical values from the Results section (e.g., specific ranges for each method across lithologies) without adequate synthesis or refinement.

Response 5: Thank you for your comment. After the reviews we recieved, we added summarizing paragraph with the overview of the applicability of the compared methods to the conclusions of the revised manuscript, which in our opinion conjoins with the rest of the conclusions into a cohesive summary. Some paragraphs may seem excesive, or repetitive, but for this kind of large coparative research we reckon that the differences need to be clearly summarized. You can find the added paragraph on page No. 24 from line No. 717 to line No. 740.

Comment 6: The language quality requires improvement; careful proofreading of grammar, tense consistency, and logic is recommended.

Response 6: Thank you for your comment. We agree that the grammar ought to be improved and unified, therefore the manuscript underwent language check and was proofread again for any discrepancies. We hope that the level of English is sufficient now.

Comments 7: Figure 3 displays a ranked-events curve, but the specific algorithmic steps for extracting baseflow values from this curve are not explained.

Response 7: Thank you for your comment and observation. We already supplemented sufficient explanation of the Killes methodology used in the research after also other reviewer pointed out this missing explanation. You can find it in the Materials and Methods section of the revised manuscript from line No. 344 page No. 11 to line No. 373 on page No. 12. A supplementary Figure No. 7 was also added in the revised manuscript, which can be found on page No. 12.

Comments 8: It is recommended to add the standard deviation or coefficient of variation for the multi-year average baseflow ratio at each station to Tables 4–6.

Response 8: Thank you for your recommendation, it is highly appreciated. In this study, the average values of baseflow for each catchment were calculated with automated software only for the whole datasets, so it is not possible to add coefficients of variation for individual hydrologic years at the moment, unfortunately. Nevertheless, thank you for your inspirating comment, we will surely look further into the multi-year average in the next phase of our research.

Comments 9: Figure 1: The watershed boundary lines and administrative boundary lines employ similar line styles, making them difficult to distinguish.

Response 9: Thank you for your comment. I do not think I fully understand it though. The line for the catchment boundary is a full line, while the administrative boundary is dashed line with color buffer. To us, this seems distinctive enough. If the catchment boundary lines should differ according to hydrogeological unit they contain, we can change the figure accordingly, but perhaps they are already distinguished in Tables 1 – 3.

Comments 10: Table 2 appears twice (on Pages 4 and 5).

Response 10: Thank you for your detailed review of the manuscript, we missed this error during our check. Based on your review, we have fixed it. The change is visible on page No. 6 line No. 189 of the revised manuscript.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors improved the paper, but not to the level required for publication. The Promethee method or another multicriteria method should be applied. Authors cannot leave research unfinished, claiming that this is not required. Research should provide the definitive answer regarding the specific ranking in the analysis situation. It cannot remain in the shape of the preliminary analysis. 

Reviewer 2 Report

Comments and Suggestions for Authors

I think that this revised version has addressed satisfactorily my comments, and that the overall quality of the manuscript has improved, and cosequently I consider that it can be published.

Reviewer 3 Report

Comments and Suggestions for Authors

The revised paper has addressed all my previous comments.